In current systems, when a radio listener indicates that they wish to obtain information about the music track (e.g. music track title and artist) that is currently playing on the radio, the radio captures a predetermined amount of audio (e.g. 10 seconds) from the music track and generates an audio fingerprint. The audio fingerprint is typically smaller than a 10 second audio sample of the music track thus reducing the amount of data that needs to be sent.
Once created, the radio sends the audio fingerprint to a music information service, such as Shazam™.
The music information service attempts to identify the music track using the audio fingerprint. For example, Shazam™ maintains a library of audio fingerprints and when Shazam™ receives an audio fingerprint it searches for a match in the library.
If the music information service is able to identify the music track using the fingerprint then information about the music track is provided to the radio. The radio may then display the information to the user and/or present the user with options for purchasing the identified music track. For example, the radio may send the information about the track to a server which provides the radio with a music library ID which can be used to purchase the music track. If the music information service is unable to identify the music track using the fingerprint then the music information service may return an error to the radio.
Each identification request costs money, takes time to complete and uses up valuable bandwidth.
The embodiments described below are not limited to implementations which solve any or all of the disadvantages of known systems.